DeepSeek-V3 vs Llama 3.1-405B

Verdict

DeepSeek-V3 leads 6–4 across 11 shared benchmarks.

DeepSeek-V3 6 · Llama 3.1-405B 4 · 1 tied · higher isn't always better — see caveats.

Per-benchmark head-to-head

DeepSeek-V3Llama 3.1-405B

93.73
ARC AI2
93.73
83.33
BBH
77.2
42.05
GPQA diamond
34.55
85.2
HellaSwag
85.6
64.85
MATH level 5
49.77
82.93
MMLU
79.33
15.75
OTIS Mock AIME 2024-2025
9.63
69.4
PIQA
71.8
2.68
SimpleBench
7.6
82.9
TriviaQA
82.7
70.4
Winogrande
78.4

A higher number is not always a better model. Each score is task performance under a disclosed harness, and some of these benchmarks are saturated or scored under mixed harnesses. Follow any benchmark to its integrity grade before reading a win as decisive.

Follow the record

Per-benchmark head-to-head

Follow the record

These models

DeepSeek-V3 vs …

Llama 3.1-405B vs …

Per-benchmark head-to-head

Follow the record

These models

DeepSeek-V3 vs …

Llama 3.1-405B vs …